Human developers should not develop with the production DB, why the hell would you give an AI the rights to touch the prod DB?

I have a similar theory about star trek. In one scene there was a blurry picture and to sharpen it Ricker said: “Computer implement recursive Algorithm”. That is equivalent to “Computer do something”. So now my theory is that there is an intelligent ship with a genius AI that carries around humans that have regressed to toddler intelligence because the AI does everything for them.

The ship is basically human daycare with lots of blinking buttons and moving pictures to keep the humans occupied while the ship does the actual (and probably boring) science.

Starfleet Academy is basically teaching them technobabble and looking great in a uniform while the AIs do the real work.

He didn’t use an IDE, but I don’t remember what he tried. He also wasn’t a fan of googling stuff. Use the man pages and nothing else.

Here is my story:

There were console outputs after nearly every line. I asked about them: “Oh, I couldn’t get the debugger to work, so I print everything to the console”

This was everywhere. The whole program was like this. On a standard Linux machine. It wasn’t even remote debugging or something. Just a local C++ program.

The filenames where written in 8+3. Again, on a modern Linux machine. His answer? “You never know where we’ll port this software to”

Onto computers that were outdated decades ago? To embedded systems? Of course he had no answer for this except “just in case…”

I could tell you more, that software was the stuff for nightmares.

Yes and no one but crypto needs that. Everyone else is much better served with traditional databases.

Blockchain solves a specific problem: safe transactions without a trusted authority.

It has a lot of downsides to solve this problem without a trusted authority, so in any case where you can use a trusted authority (for example a central server) it’s much better to use that instead of a blockchain.

So everyone who added blockchains to their projects gained all the downsides while never having the problem it was meant to solve in the first place.

AIs, and I assume you meant LLMs with that, are a different breed. LLMs are new: never before could a computer handle natural language to such a degree.

Problem is, that it’s still new. So no one knows what the “killer applications” are or what monetization should look like, or what the laws about it will be.

People just throw every against the wall and see what sticks… And hope for AGI/ASI and to be on the side that rides that nearly infinite potential to the moon.

Or, you know, crash and burn in case AI reaches a wall/diminishing returns/systemic problems that can’t be fixed.

We will see.

Ah, I simply avoid that problem by being neither cute nor hot - and therefore single.

Oh yeah, I overlooked that detail.

What the hell? If they just asked ChatGPT it would have been much better:

• Repository: A folder of project files on the internet.

• Fork: Your own copy of that folder to make changes.

• Pull Request: A way to ask, “Can my changes go back into the original folder?”

All programs can be written with on less line of code. All programs have at least one bug.

By the logical consequences of these axioms every program can be reduced to one line of code - that doesn’t work.

One day AI will get there.